Liang Wang

Current knowledge graph reasoning methods actually use the same state representations and a fixed decision method to complete the total reasoning process. They are not able to truly use the different levels of attribute information. To overcome this problem, we propose a multi-level reinforcement learning reasoning model based on adaptive hierarchical feature fusion. Our model divides attribute information into coarse-grained attributes and fine-grained attributes. In this way, our model can better balance overall semantics and details, thereby holding connections between entities and relationships. We design an adaptive hierarchical feature fusion mechanism. Our model dynamically determines whether it pays more attention to attribute information or relationship according to the needs of the current task, rather than constantly using the same weights. This enables our model to have better performance in different reasoning situations. We improve the decision process of the model, making the model know which relationships to focus on and which are less relevant to skip, thereby reducing ineffective search. According to the results of the experiment, our method is superior to the existing reinforcement learning reasoning methods, including MINERVA, AdaPath, CURL, and HMR. Our model improves MRR, Hits@1, Hits@3, and Hits@10 indicators compared with these baseline models. This shows that our method can improve the accuracy of inference and verify the effectiveness of our model in complex knowledge graph inference tasks.

Extracting structured threat intelligence from unstructured cybersecurity texts requires accurate identification of entities together with their underlying semantic relations. However, threat reports often exhibit intricate sentence structures, long-range contextual dependencies, and tightly coupled entity–relation patterns, which pose substantial challenges for existing extraction approaches. To address these challenges, this study investigates joint entity–relation extraction from the perspective of semantic dependency modeling and develops HDIM-JER, a unified framework that captures structured interactions among heterogeneous linguistic features. HDIM-JER integrates character-level cues, contextual representations, and higher-order semantic dependency evidence to enhance structural awareness during joint inference, where different second-order dependency configurations provide an interpretable perspective on structurally symmetric and hierarchically asymmetric interaction patterns among entity–relation instances. By incorporating multi-level dependency interactions, HDIM-JER effectively alleviates error propagation associated with pipeline-based architectures and improves the modeling of complex relational dependencies. Extensive experiments on a threat intelligence corpus and a public benchmark dataset demonstrate consistent performance improvements over representative state-of-the-art methods in both entity recognition and relation extraction, confirming the effectiveness of higher-order semantic dependency interaction modeling for threat intelligence analysis.

The symmetry between different representation spaces plays a crucial role in effectively modeling complex multimodal data. To address the challenge of equipment knowledge graphs containing hierarchical relationships that cannot be fully represented in a single space, this study proposes UMEAD, an unsupervised multimodal entity alignment method based on dual-space embeddings. The method simultaneously learns graph embeddings in both Euclidean and hyperbolic spaces, forming a structural symmetry where the Euclidean space captures local regularities and the hyperbolic space models global hierarchies. Their complementarity achieves a balanced and symmetric representation of multimodal knowledge. An adaptive feature fusion strategy is further employed to dynamically weight semantic and visual modalities, enhancing the symmetry and complementarity between different modalities. To reduce reliance on scarce pre-aligned data, pseudo seed instances are generated from multimodal features, and an iterative constraint mechanism progressively enlarges the training set, enabling unsupervised alignment. Experiments on public datasets, including EMMEAD, FB15K-DB15K, and FB15K-YAGO15K, demonstrate that the combination of dual-space embeddings, adaptive fusion, and iterative constraints significantly improves alignment accuracy. In summary, the proposed method reduces dependence on pre-aligned data, strengthens multimodal and structural alignment, and its symmetric embedding and fusion design offers a promising approach for the construction and application of multimodal knowledge graphs in the equipment domain.

Image segmentation is an important method in the field of image processing, while infrared (IR) image segmentation is one of the challenges in this field due to the unique characteristics of IR data. Infrared imaging utilizes the infrared radiation emitted by objects to produce images, which can supplement the performance of visible-light images under adverse lighting conditions to some extent. However, the low spatial resolution and limited texture details in IR images hinder the achievement of high-precision segmentation. To address these issues, an attention mechanism based on symmetrical cross-channel interaction—motivated by symmetry principles in computer vision—was integrated into a Mask Region-Based Convolutional Neural Network (Mask R-CNN) framework. A Bottleneck-enhanced Squeeze-and-Attention (BNSA) module was incorporated into the backbone network, and novel loss functions were designed for both the bounding box (Bbox) regression and mask prediction branches to enhance segmentation performance. Furthermore, a dedicated infrared image dataset was constructed to validate the proposed method. The experimental results demonstrate that the optimized model achieves higher segmentation accuracy and better segmentation performance compared to the original network and other mainstream segmentation models on our dataset, demonstrating how symmetrical design principles can effectively improve complex vision tasks.

Key element extraction is an important research field in computational linguistics. Based on the Hidden Markov Model in natural language processing technology, this paper utilizes the Viterbi decoding algorithm, along with its optimization and improvement algorithms, to construct a key element extraction model. This model then extracts the key legal elements from the original corpus of traffic collision litigation cases. To further validate the performance of this paper’s model, a public newspaper dataset released by a university was selected to deeply explore its effectiveness. This paper’s model significantly improves the accuracy and F1 values of the key elements of legal text extraction, reaching 93.28% and 90.83%, respectively, compared to all other models. The model’s extraction effect on the six key elements in the legal text reaches an ideal state, where the F1 value of the extracted element’s sentence results reaches 100%. In comparison to the HMM model in the public dataset, the model in this paper has improved by 10.93%, 8.78%, and 10.29% in the three indexes, indicating its superior performance in key element extraction.